Birth seasonality in bipolar disorder, schizophrenia, schizoaffective disorder and stillbirths.
BACKGROUND: More than 40 studies have been done on seasonal birth patterns for schizophrenia, but only two small studies have been done for DSM-III-R bipolar disorder and none for schizoaffective disorder. Two studies have also reported a significant relationship between schizophrenia births and stillbirths. METHODS: In the largest study to date, birth data from four states was obtained on 126,987 state psychiatric hospitals inpatients divided into 'process' schizophrenia (disorganized, catatonic, undifferentiated), paranoid schizophrenia, schizoaffective disorder, bipolar disorder and major depression. Time series analysis compared these births to all general births and to stillbirths. RESULTS: 'Process' schizophrenia, paranoid schizophrenia, schizoaffective disorder and bipolar disorder all had statistically significant seasonal excess births from December through March (p = 0.0000). The largest excess was 5.8% for bipolar disorder. Major depression had significant excess births from March through May. Time series analysis showed statistically significant coherences between major depression and bipolar disorder (0.995) and between schizoaffective disorder and both 'process' schizophrenia (0.977) and bipolar disorder (0.977). Unexpectedly, a significant coherence was also found between paranoid schizophrenia and bipolar disorder (0.972). Excess stillbirths were found for each month from January through June and a significant coherence was found between stillbirths and paranoid schizophrenia (0.998). CONCLUSIONS: This study demonstrates that DSM-III-R bipolar disorder and schizoaffective disorder both have an excess of winter births, similar to that found in schizophrenia. Time series analysis, however, suggests that the causes may not be identical. Major depression, by contrast, has an excess of spring births.